INDEX

Explanations

film and event screenings

The neuron strongly activates on occurrences of “screening” (and its forms like “screened” or “screenings”) in the text.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

Maz

-0.90

 PLOS

-0.88

fclose

-0.75

alkin

-0.73

 PLoS

-0.73

esac

-0.71

 aufgenommen

-0.71

 vocal

-0.71

Deserializer

-0.71

yttää

-0.71

POSITIVE LOGITS

 screening

5.22

 screenings

4.38

screening

4.19

 Screening

4.09

Screening

4.09

 screened

3.64

 showing

2.69

 screen

2.52

 SCREEN

2.47

 screens

2.31

Activations Density 0.076%